Effects of aerobic exercise performed in fasted v. fed state on fat andcarbohydrate metabolism in adults: a systematic review and meta-analysis
Alexandra Ferreira Vieira2*, Rochelle Rocha Costa1,2, Rodrigo Cauduro Oliveira Macedo1,3,Leandro Coconcelli1,2 and Luiz Fernando Martins Kruel1,2
1Physical Education, Physiotherapy and Dance School, Federal University of Rio Grande do Sul, 750, Felizardo Street,90690-200 Porto Alegre, Brazil2Research Group on Water and Land Activities, Federal University of Rio Grande do Sul, 750, Felizardo Street,90690-200 Porto Alegre, Brazil3Research Group on Exercise Physiology and Biochemistry, Federal University of Rio Grande do Sul, 750, Felizardo Street,90690-200 Porto Alegre, Brazil
(Submitted 7 March 2016 – Final revision received 29 July 2016 – Accepted 3 August 2016 – First published online 9 September 2016)
AbstractThis study aimed to verify the effect of aerobic exercise performed in the fasted v. fed states on fat and carbohydrate metabolism in adults.Searches were conducted in March 2015, and updated in July 2016, using PubMed®, Scopus and Cochrane databases (terms: ‘fasting’,‘exercise’, ‘aerobic exercise’, ‘substrate’, ‘energy metabolism’, ‘fat’, ‘glucose’, ‘insulin’ and ‘adult’) and references from selected studies. Trialsthat compared the metabolic effects of aerobic exercise (duration ≤120min) performed in the fasted v. fed states in adults were accepted. Theoutcomes evaluated were fat oxidation during exercise and the plasma concentrations of insulin, glucose and NEFA before and immediatelyafter exercise; two independent reviewers extracted the data (A. F. V. and L. C.). The results were presented as weighted mean differencesbetween treatments, with 95% CI. Of 10 405 articles identified, twenty-seven studies – with a total of 273 participants – were included. Therewas a significant increase in fat oxidation during exercise performed in the fasted, compared with fed, state (−3·08 g; 95% CI −5·38, −0·79; I2
39·1%). The weighted mean difference of NEFA concentrations was not significantly different between states (0·00mmol/l; 95% CI −0·07,0·08; I 2 72·7%). However, the weighted mean differences of glucose (0·78mmol/l; 95% CI 0·43, 1·14; I 2 90·8%) and insulin concentrations(104·5 pmol/l; 95% CI 70·8, 138·2; I 2 94·5%) were significantly higher for exercise performed in the fed state. We conclude that aerobicexercise performed in the fasted state induces higher fat oxidation than exercise performed in the fed state.
Key words: Fasting: Exercise: Energy metabolism: Reviews
Fasting is characterised by the absence of food and/or energybeverage intake for a period of time, which may last from severalhours to a few weeks(1,2). However, most people fast for 8–12hdaily – the ‘overnight fasting’ period(2). During this period, NEFA,ketone bodies and glucose derived from liver glycogen andgluconeogenesis are the predominant energy sources(3).During exercise, NEFA also make a considerable contribution
to energy metabolism owing to the increased availability ofthese substrates in the plasma. This is caused by increasedadrenaline levels and decreased insulin concentrations in theblood(4). Fasting promotes low levels of insulin and hepaticglycogen(2). Thus, when aerobic exercise is performed underthese conditions, an increase in the utilisation of fat as anenergy substrate is observed, when compared with exerciseperformed in the fed state(5,6). The decrease in fat oxidationduring exercise in the fed state can be mainly attributed tohigher insulin concentrations caused by a meal, which may
inhibit the breakdown of intramuscular TAG (IMTG) andreduce the availability of NEFA for oxidation(7,8).
Several studies have indicated that regular exercise promotesbeneficial effects in terms of health and body composition(9–11),including an improvement in insulin sensitivity and main-tenance and reduction of body weight and body fat. It has beensuggested that exercise enhances fat oxidation and thatthis adaptation may be associated with improved insulinsensitivity(12). Furthermore, higher fat oxidation capacity duringexercise seems to be related to a decrease in the number ofmetabolic risk factors(13). Venables & Jeukendrup(14) demon-strated that participating in a training programme for 4 weeks,with continuous aerobic exercise programmed for the max-imum contribution of fat as the energy substrate during eachsession, can further increase fat oxidation. This higher oxidationwas associated with improvements in insulin sensitivity inobese men. In healthy, young men, the maximal fat oxidation
Abbreviation: IMTG; intramuscular TAG.
* Corresponding author: A. F. Vieira, fax +55 51 3308 5820, email [email protected]
during exercise was positively associated with insulin sensitivityand 24-h fat oxidation(15). Studies have demonstrated thatexercise performed in the fasted state can increase the rate of fatoxidation at rest from 9(16) to 24 h(17–19) after exercise whencompared with the same exercise performed after a meal. Thishigher utilisation of fat as an energy source at rest may promotereduction in body fat.On the basis of these data, aerobic exercise performed in the
fasted state has been considered a strategy to increase fatoxidation during exercise and, chronically, to promote adapta-tions that may be beneficial to health. However, although moststudies reported higher fat oxidation under these conditionscompared with a carbohydrate-fed state, it is not clear whetherthe stimulation of lipolytic activity and/or decreasedre-esterification of NEFA that occur during the fasted state(20)
result in a significantly increased use of fat as an energy substrateduring exercise. This systematic review with meta-analysis aimedto verify the effect of aerobic exercise performed during fasted v.fed states on fat and carbohydrate metabolism in adults.
Methods
Eligibility criteria
This review considered clinical trials (parallel and randomisedcross-over designs) evaluating the effect of performing anaerobic exercise intervention of no >120min duration (or dataat 120min for those interventions with longer durations) ina fasted state among adults aged 19–59 years. These interven-tions had to be compared with the same exercise performed inthe fed state (prior consumption of meals containing at least25 g of carbohydrates)(21). Included studies evaluated thefollowing outcome measures: fat oxidation during exercise,considered the primary end point; and serum concentrations ofNEFA, glucose and insulin immediately before and after theexercise session; the absolute weighted mean differences ofthese concentrations were considered the secondary endpoints. Studies that evaluated these acute responses to aerobicexercise were included; trials that did not present acuteoutcome data were excluded. In the case of trials with severalpublications (or sub-studies), the study was included only once.
Search strategy
The electronic databases MEDLINE® (via PubMed®), Scopusand Cochrane were used. In addition, manual searches wereconducted of references of studies identified for inclusion. Thesearch was conducted in March 2015 and updated in July 2016.The terms ‘fasting’, ‘exercise’, ‘aerobic exercise’, ‘substrate’,‘energy metabolism’, ‘fat’, ‘glucose’, ‘insulin’ and ‘adult’ as wellas related entry terms were used. The searches were limitedto articles published in English, Portuguese and Spanishlanguages. The search strategy used in the PubMed® database isavailable as the online Supplementary Material. Details of otherstrategies may be obtained upon request. This systematicreview and meta-analysis was prepared and is presented inaccordance with ‘Preferred Reporting Items for SystematicReviews and Meta-Analyses’ guidelines(22,23).
Selection of studies
Selection of studies for review was performed independently andduplicated, without restriction on the date of publication. First,the titles and abstracts of all articles identified by the searchstrategy were evaluated for inclusion independently by tworesearchers (A. F. V. and L. C.), in duplicate form. Whenever theabstract did not provide sufficient information about inclusionand exclusion criteria, the full article was evaluated. Second, thesame reviewers independently evaluated the full articles of thoseidentified as appropriate from the abstract screening process, andmade their selection according to eligibility criteria. Disagree-ments between reviewers were resolved by consensus, and inthe case of continuing disagreement the evaluation was made bya third reviewer (R. R. C.). To avoid possible double counting ofparticipants included in more than one report by the sameauthors/working groups, the periods of recruitment of partici-pants and areas of recruitment were evaluated, and authors werecontacted for clarification where necessary.
Data extraction
Data extraction was performed by two reviewers (A. F. V. andL. C.) independently concerning methodological characteristics,interventions and outcomes of the studies using a standardisedform. As in the selection stage, disagreements were resolvedby consensus or by a third reviewer (R. R. C.). The extracteddata included average age, BMI, sex and training status ofparticipants; exercise duration and intensity; time between dietaryintake and the start of exercise; amount of carbohydrateconsumed in the pre-exercise meal; and the end points analysed.If the required data were not found in the published report, thecorresponding author was contacted to provide missing data and,in the absence of responses or data extraction alternatives, thestudy or missing end point was excluded from the review. Datapresented only graphically, and for which more detail was notprovided despite a request to the corresponding authors, wereextracted using ‘DigitizeIt’ software. Where it was not possible toextract means or standard deviations from graphs at the requiredpoints, the variable was excluded from the analysis.
In this phase, studies that included diabetic participants, or thosein which carbohydrates were provided during exercise as part ofthe study protocol, were excluded to avoid possible bias in theresults. The primary end point we assessed was the total absoluteaverage fat oxidation during exercise. Secondary end points werethe weighted mean difference in insulin, glucose and NEFA con-centrations. Weighted mean differences were calculated fromvalues taken immediately before and during the last minute ofexercise for studies lasting ≤120min. For studies of longer dura-tion, the time ‘120min’ was considered the last minute of exercise.
In studies where the total absolute average fat oxidation duringexercise was not presented in the published article, a request wassubmitted to the authors, and if means for VO2 and carbondioxide production values were provided these were applied tothe formula determined by Péronnet & Massicotte(24) in order todetermine the fat oxidation rate. The units of measurementsused in this review were grams for fat oxidation, mmol/l forconcentrations of NEFA and glucose, and pmol/l for insulin
1154 A. F. Vieira et al.
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
concentrations. Study data not presented in these units wereconverted. For instance, where fat oxidation was presented usingan energy value (kJ/kcal), these averages were divided by 40·79kJ(9·75kcal) in order to obtain the value in grams(25). If these datawere not provided by authors, or if it was not possible to calculatethe total oxidised absolute average during exercise, the variable orthe study was excluded. Studies with two or more comparisongroups with the same population were included with only onecomparator, which was selected according to the time betweendietary intake and exercise and/or the nutritional characteristics ofmeals consumed that most closely resembled the other studiesbeing reviewed, in an effort to standardise results. For studies withtwo or more intervention groups, a single group was also inclu-ded, selected according to characteristics similar to other studies.
Evaluation of risk of bias
The assessment of the methodological quality of includedstudies was performed according to criteria proposed byCochrane(26): appropriate use of randomisation sequences,allocation concealment, blinding of participants and/or thera-pists, blinding of assessors to outcomes, and description oflosses and exclusions. When these processes had been descri-bed in the published document, it was considered that criteriahad been met and these studies were classified as being at ‘lowrisk’ of bias and, in opposition, as ‘high risk’. Studies that did notreport these data were classified as ‘unclear risk’. Descriptionsof losses and exclusions were considered ‘low risk’ when thenumber of participants evaluated were presented in the legendsof charts and graphs. Quality evaluation was performedindependently by two reviewers (A. F. V. and L. C.).
Data analysis
Results are presented as weighted mean differences for absolutevalues between treatments with 95% CI. The standard deviationof mean difference values not provided by studies was imputedaccording to the equation proposed by Higgins et al.(27).Statistical heterogeneity of treatment effects between studieswas evaluated by Cochran’s Q test and I 2 inconsistency test;values above 50% indicated high heterogeneity(28). In case oflow heterogeneity, the fixed effect model was used to poolstudy results for the outcomes. When significant heterogeneitywas observed (I 2> 50%), the random effects model wasapplied. Meta-analyses comprised comparisons of fat oxidationduring aerobic exercise performed in the fasted v. fed state andthe changes in concentrations (expressed using weighted meandifferences) of glucose, NEFA and insulin from immediatelybefore exercise to the last minute of exercise (post-exercise).Values of α≤ 0·05 were considered statistically significant.For variables with high heterogeneity, sensitivity analyses
were performed according to the following criteria: exercisetime, exercise intensity, sex of participants, BMI of participants,training level of participants, pre-exercise values for each vari-able, time between dietary intake and the start of exercise, andamount of carbohydrate consumed in the pre-exercise meal.Furthermore, publication bias was assessed using funnel plots
for each outcome (of each trial’s effect size against the standard
error). Funnel plot asymmetry was evaluated using Begg andEgger tests(29), and a significant publication bias was consideredif P< 0·10. The trim-and-fill computation was used to estimatethe effect of publication bias on the interpretation of results.
All analyses were performed using ComprehensiveMeta-Analysis version 2.0, except the risk of bias, which wasperformed using Review Manager version 5.3 (CochraneCollaboration).
Results
Description of studies
Of the 10 405 studies identified from the database searches,twenty-three met our inclusion criteria. An additional four studieswere included from a manual search of the reference lists of theincluded studies, bringing the total number of articles included totwenty-seven. Of these, three studies(30–32) were included twicebecause they had met eligibility criteria for two groups withdifferent populations, in which each population had a differentintervention group and control group: references ‘Bergman &Brooks, 1999a’ and ‘Montain et al., 1991a’ related to populationscomprised of trained men, ‘Bergman & Brooks, 1999b’ and‘Montain et al., 1991b’ related to populations comprised ofuntrained men, and ‘Isacco et al., 2012a’ and ‘Isacco et al., 2012b’related to populations of women who did not and did use thecontraceptive pill, respectively. Thus, thirty comparisons wereused in this meta-analysis (Fig. 1). In total, 270 and 269 partici-pants were included in the fasted and fed groups, respectively.The majority of studies (80%) analysed men, whereas 13·3%analysed women, and 6·6% analysed both sexes. Most samplescomprised physically active individuals (86·7%), and exercisesessions lasted an average of 73min. The meals were provided30–240min before the interventions and were composed of amaximum of 215 g of carbohydrates (Table 1).
In all, four studies were excluded from our analysis: one studywas unaccessible(33), and the other three met all eligibility criteria,but were not used because of the unavailability of results(34),or the presentation of averages(19) and standard deviations(35)
graphically, with no clarification received from authors and nopossibility of data extraction using ‘DigitizeIt’ software.
From some of the studies included, it was necessary to excludecertain variables because absolute averages were not given – forexample, the absolute average of fat oxidation duringexercise(6,7,36–38). Other variables were excluded as it wasimpossible to extract values for standard deviation of insu-lin(39–42), glucose(40,41,43) and NEFA(32,40) concentrations. Thesedata were all requested from authors but were not provided.‘DigitizeIt’ software was used to extract the average relating to fatoxidation from one study(44), relating to NEFA concentrationsfrom fourteen studies(6,7,37,38,41–43,45–51), relating to glucose con-centrations from sixteen studies(6,7,32,36–39,42,45–52) and relating toinsulin concentrations from thirteen studies(6,7,37,38,43,45–52).
Risk of bias
Of the included studies, 80% showed adequate generationof randomisation sequence, 6·6% reported allocation
Exercise in fasted adults and fat oxidation 1155
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
concealment, 20% had blinded participants and/or therapists,6·6% had blinded the assessors to the outcomes and 66·6%described losses to follow-up and exclusions (Fig. 2 and 3).
Effects of interventions
Fat oxidation. Data on fat oxidation were available from elevenstudies(30,31,39–41,43,44,47,50–52), with a total of 117 individualsevaluated (Fig. 4). Aerobic exercise performed in the fastedstate was associated with a significant increase in fat oxidationduring exercise when compared with the fed state (effect size:−3·53; 95% CI −4·76, −2·30; I 2 39·1%). Aerobic exerciseperformed in the fasted state led to an increase in fat oxidationof approximately 3·53 g, compared with execution of the sameexercise after consumption of meals containing carbohydrates.However, the analysis of publication bias identified a significantbias (P= 0·007), and thus the adjusted value of the effect size,according to the Duval & Tweedie’s trim and fill test, resultedin 3·08 g.Given the influence of exercise intensity on fat oxidation,
sensitivity analyses were performed to identify whether therewas an effect difference when stratified by two differentintensity ratings – VO2max <70% and VO2max ≥70%. Thus, eventhough the meta-analysis did not demonstrate significant het-erogeneity (P= 0·07), sensitivity analyses were performed:<70% VO2max (3·45 g; 95% CI 2·19, 4·71; P< 0·001; I 2 50%) and≥70% VO2max (5·38 g; 95% CI −0·45, 11·21; P= 0·07; I 2 0%).Aerobic exercise of low-to-moderate intensity performed in thefasted state induced a higher fat oxidation compared with a fed
state. On the other hand, there was no significant differencebetween fasted and fed states in relation to fat oxidation duringaerobic exercise of moderate-to-high intensity.
Sensitivity analyses were also performed for fat oxidationtaking the following into account: exercise time (≤60min:3·35 g; 95% CI 2·07, 4·62; P< 0·001; I 2 54%; >60min: 6·13 g;95% CI 1·37, 10·88; P= 0·01; I 2 9%); sex of participants (male:6·39 g; 95% CI 3·84, 8·94; P< 0·001; I 2 0%; female: 2·60 g;95% CI 1·19, 4·01; P= 0·0003; I 2 0%); BMI of participants(<25 kg/m2: 2·79 g; 95% CI 1·42, 4·17; P< 0·001; I 2 30%);training level of participants (physically active: 3·74 g; 95% CI1·97, 5·52; P< 0·001; I 2 49%; sedentary: 3·34 g; 95% CI 1·62,5·05; P= 0·0001; I 2 23%); time between consumption of mealand the beginning of exercise (<100min: 3·41 g; 95% CI 1·68,5·14; P= 0·0001; I 2 57%; >100min: 3·66 g; 95% CI 1·91, 5·41;P< 0·001; I 2 37%); and quantity of carbohydrate consumed inthe pre-exercise meal (<100 g: 3·51 g; 95% CI 1·84, 5·17;P< 0·001; I 2 34%; ≥100 g: 3·56 g; 95% CI 1·73, 5·39; P= 0·0001;I 2 53%). Thereby, these results demonstrated no change inthe pattern already presented, of higher fat oxidation whenthe exercise is performed in the fasted state, regardless of theadopted criteria for the sensitivity analyses.
NEFA. Data on NEFA concentrations were available fromsixteen studies(6,7,31,37,38,41–43,45–52), with a total of 144 indivi-duals evaluated (Fig. 5). All but one of these studies usedthe same sample populations for both interventions(37). Theweighted mean difference of NEFA before and after exercise
10 405 articles identified throughdatabase search
Iden
tific
atio
n
9857 articles after the removal ofduplicates
69 full-text articles assessed for eligibility
27 articles included 3 studies with two different populations
4 studies selected based on analysis of thereferences of included studies
46 exclusions after reading the full-text
4 did not study the target population
3 did not evaluate the interest variables
23 did not perform the intervention10 did not have the adequate comparator
2 studies with multiple publications4 studies without access to the results
9788 exclusions based on the title and/orsummary review
2073 did not study the target population5020 did not perform the intervention40 did not have the adequate comparator56 did not evaluate the interest variables2599 did not have the appropriate design
Sel
ectio
nE
ligib
ility
Incl
usio
n
Fig. 1. Flow chart of the included studies.
1156 A. F. Vieira et al.
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
Table 1. Characteristics of included studies(Mean values and standard deviations)*
Age (years)Exercise duration Time between meal Amount of carbohydrate
Studies Mean SD Sex/n Training status (min) Exercise intensity and exercise (min) pre exercise meal (g)
Aziz et al.(53) 27·3 7·2 Male/10 Physically active 60 65·0% VO2max 180 to 240 126·7Bergman & Brooks(30) 25·1 4·8 Male/7 Physically active 90 60·0% VO2peak 180 119·6Bergman & Brooks(30) 22·1 3·4 Male/7 Sedentary 120 40·0% VO2peak 180 119·6Bouhlel et al.(54) 19·0 2·0 Male/9 Physically active 30 20·0, 30·0, 40·0, 50·0, 60·0% Wmax Does not mention Does not mentionCoyle et al.(45) 25·0 5·0 Male/7 Physically active 105 70·0% VO2max 240 141·8Coyle et al.(7) 22·0 4·9 Male/6 Physically active 40 50·0% VO2max 60 and 10 96·6Dohm et al.(46) 28·7 3·9 Male/9 Physically active 90 or until exhaustion
(about 80)70·0–75·0% VO2max 120 to 240 47·0
Farah & Gill(39) 28·1 10·7 Male/10 Sedentary 60 50·0% VO2max 30 56·5Gonzalez et al.(47) 23·2 4·3 Male/12 Physically active 59 61·1% VO2peak 120 66·6Guéye et al.(55) 22·5 1·7 Male/12 Physically active 60 75·0% HRmax Does not mention Does not mentionHorowitz et al.(6) 26·5 9·3 Male/6 Physically active 60 44·0% VO2peak 60 60·0Isacco et al.(31) 22·9 3·6 Female/10 Sedentary 45 65·0% VO2max 180 72·2Isacco et al.(31) 21·2 1·9 Female/11 Sedentary 45 65·0% VO2max 180 73·4Kirwan et al.(48) 22·0 2·4 Male/6 Physically active Until exhaustion (120) 60·0% VO2peak 45 75·0Kirwan et al.(49) 24·0 4·9 Female/6 Physically active Until exhaustion (120) 60·0% VO2peak 45 75·0Little et al.(36) 23·3 3·8 Male/7 Physically active 90 (45) Vmax 180 86·0Little et al.(50) 22·8 3·2 Male/13 Physically active 105 Vmax 120 Does not mention (1·5 g/kg)Massicotte et al.(52) 24·8 (SD 6·9) (fast)
34·0 8·9 Male/5 Physically active Until exhaustion (90) 70·0% VO2max 45 69·8
Montain et al.(32) Does not mention Male/9 Physically active 30 70·0% VO2peak 120 131·6Montain et al.(32) Does not mention Male/8 Physically active 30 70·0% VO2peak 120 154·6Paul et al.(41) 24·9 3·4 Mixed/12 Physically active 90 60·0% VO2peak 90 32·4Ramos-Jiménez
et al.(56)22·5 3·7 Mixed/30 Physically active 8 to 15 98·0% HRmax 70 Does not mention
(60% ETV meal)Satabin et al.(42) 25·2 17·7 Male/9 Physically active 110 60·0% VO2max 60 100·0Schabort et al.(37) 26·0 7·9 Male/7 Physically active 105 70·0% VO2max 180 100·0Shin et al.(38) 23·3 2·5 Male/8 Physically active 60 50·0% VO2max 30 66·4Whitley et al.(43) 21·0 10·8 Male/8 Physically active 90 70·0% VO2max 240 215·0Willcutts et al.(44) 23·7 2·4 Female/8 Physically active 30 (23) 62·0% VO2max 90 109·3Wu et al.(51) 26·8 3·3 Male/9 Physically active 60 65·0% VO2max 180 141·0Ziogas & Thomas(57) 27·4 3·8 Male/7 Physically active 60 60·0% VO2max 180 111·5
VO2peak, peak VO2; Wmax, maximum power; HRmax, maximum heart rate; Vmax, maximum velocity; ETV, energy total value.* Exercise duration: total time of exercise duration evaluated in the study (time post exercise extracted).
Exercise
infasted
adults
andfat
oxid
ation
1157
Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 23 May 2020 at 01:01:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114516003160
was not demonstrated to be significantly different betweenexercise performed in the fasted or fed states (effect size: 0·00;95% CI −0·07, 0·08; I 2 72·7%). The analysis of publication biasfor this outcome showed no significant bias (P= 0·124).Owing to the high heterogeneity (P< 0·001) found in the
analysis of this variable, sensitivity analyses were performed.Significant heterogeneity was found for most of the variablesanalysed: exercise time (≤60min: I 2 93%; P< 0·001; >60min:I 2 83%; P< 0·001); exercise intensity (<70% VO2max: I
2 92%;P< 0·001; ≥70% VO2max: I
2 74%; P= 0·004); sex of participants(male: I 2 90%; P< 0·001; female: I 2 90%; P< 0·001); BMI ofparticipants (<25 kg/m2: I 2 90%; P< 0·001; >25 kg/m2: I 2 0%;P= 0·76); training level of participants (physically active:I 2 90%; P< 0·001; sedentary: I 2 0%; P= 0·65); pre-exercisevalues for fasting (<1mmol/l: I 2 90%; P< 0·001); time betweenconsumption of meal and the beginning of exercise (<100min:I 2 85%; P< 0·001; >100min: I 2 82%; P< 0·001); and quantity ofcarbohydrate consumed in the pre-exercise meal (<100 g:I 2 87%; P< 0·001; ≥100 g: I 2 90%; P< 0·001). Sensitivity ana-lyses for the criterion ‘pre-exercise values in fasting >1mmol/l’were not performed because only one study presented thischaracteristic. The criteria ‘BMI> 25 kg/m2
’ and ‘sedentary’showed no significant heterogeneity, although sensitivity ana-lyses were performed with only two studies for each. Owing tothe maintenance of high heterogeneity and/or low numberof studies, the data presented graphically (Fig. 5) refer to thegeneral analysis (disregarding the sensitivity analysis).
Glucose. Data on glucose concentrations were availablefrom twenty-two studies(6,7,31,32,36–39,42,45–57), with a total of226 individuals evaluated (Fig. 6). All but one of these studiesused the same sample populations for both interventions(37).Significantly lower variation was reported for glucose con-centrations from before to after exercise in the fasted v.fed states (effect size: 0·60; 95% CI 0·25, 0·94; I 2 90·8%).Nevertheless, the analysis of publication bias identified asignificant bias (P= 0·057), and thus the adjusted value of theeffect size, according to the Duval & Tweedie’s trim-and-fill test,resulted in 0·78mmol/l.Because of the high heterogeneity (P= 0·001) found for
this variable, sensitivity analyses were performed and, again,significant heterogeneity was found in most analyses. Thesevariables were as follows: exercise time (≤60min: I 2 95%;P< 0·001; >60min: I 2 99%; P< 0·001); exercise intensity(<70% VO2max: I 2 98%; P< 0·001; ≥70% VO2max: I 2 80%;
P< 0·001); sex of participants (male: I 2 95%; P< 0·001; female:I 2 99%; P< 0·001); BMI of participants (<25 kg/m2: I 2 99%;P< 0·001; >25 kg/m2: I 2 82%; P= 0·001); training level ofparticipants (physically active: I 2 97%; P< 0·001; sedentary:I 2 16%; P= 0·30); pre-exercise values in fasting (<5mmol/l:I 2 95%; P< 0·001; >5mmol/l: I 2 99%; P< 0·001); time betweenconsumption of meal and the beginning of exercise (<100min:I 2 98%; P< 0·001; >100min: I 2 69%; P< 0·001); and quantity ofcarbohydrate consumed in the pre-exercise meal (<100 g:I 2 98%; P< 0·001; ≥100 g: I 2 44%; P= 0·08). Again, the criterion‘sedentary’ showed no significant heterogeneity, although thesensitivity analysis was performed with only two studies. Thecriterion ‘quantity of carbohydrate consumed in the pre-exercise meal ≥100 g’ was analysed with eight interventions(n 66), and did not show significant heterogeneity. Therefore, inthis case, the weighted mean difference of relative glucoseconcentrations did not appear to differ significantly whenexercise was performed in a fasted v. fed state (P= 0·91).Because of the high heterogeneity and/or low number of stu-dies, the data presented graphically (Fig. 6) refer to the generalanalysis (disregarding the sensitivity analysis). More detailedresults of the sensitivity analysis performed for this variablerelating to the criterion ‘quantity of carbohydrate consumed inthe pre-exercise meal ≥100 g’ can be provided on request.
Insulin. Data on insulin concentrations were available fromfifteen studies(6,7,31,32,37,38,43,45–52), with a total of 140 individualsevaluated (Fig. 7). Again, all but one of these studies usedthe same sample populations for both interventions(37).Significantly lower variation was reported for insulin con-centrations from before to after exercise in the fasted v.fed states (effect size: 104·5; 95% CI 70·8, 138·2; I 2 92·5%).However, the analysis of publication bias identified a significantbias (P< 0·001), and thus the adjusted value of the effect size,according to the Duval & Tweedie’s trim and fill test, resultedin 104·5 pmol/l.
As with the other blood variables, high heterogeneity wasfound (P< 0·001) and sensitivity analyses were consequentlyperformed. Once again, significant heterogeneity was found formost comparisons: exercise time (≤60min: I 2 82%; P< 0·001;>60min: I 2 95%; P< 0·001); exercise intensity (<70% VO2max:I 2 91%; P< 0·001; ≥70% VO2max: I
2 89%; P< 0·001); sex ofparticipants (male: I 2 93%; P< 0·001; female: I 2 96%;P< 0·001); BMI of participants (<25 kg/m2: I 2 90%; P< 0·001;>25 kg/m2: I 2 98%; P< 0·001); training level of participants
Random sequence generation (selection bias)
Allocation concealment (selection bias)
Blinding of participants and personnel (performance bias)
Blinding of outcome assessment (detection bias)
Incomplete outcome data (attrition bias)
0 25 50 75 100%
Fig. 2. Risk of bias in the included studies. , Low risk of bias; , unclear risk of bias; , high risk of bias.
1158 A. F. Vieira et al.
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
(physically active: I 2 95%; P< 0·001; sedentary: I 2 0%;P= 0·39); pre-exercise values in the fed state (<200 pmol/l:I 2 92%; P< 0·001; >200pmol/l: I 2 83%; P< 0·001); time betweenconsumption of meal and the beginning of exercise (<100min:I 2 95%; P< 0·001; >100min: I 2 91%; P< 0·001); and quantity ofcarbohydrate consumed in the pre-exercise meal (<100g:I 2 91%; P< 0·001; ≥100g: I 2 92%; P< 0·001). Once again, thevariable ‘sedentary’ showed no significant heterogeneity,although the sensitivity analysis was performed with only twointerventions. Owing to the high heterogeneity and/or lownumber of studies, the data presented graphically (Fig. 7) refer tothe general analysis (disregarding the sensitivity analysis).
Discussion
The major finding of this systematic review with meta-analysis isthat performing aerobic exercise at low-to-moderate intensity inthe fasted state induces a significant increase (3·08 g) in fatoxidation while exercise is being performed. No difference wasseen in the variation of NEFA concentrations between exerciseperformed in the fasted v. fed states. However, greater varia-tions in glucose and insulin concentrations were seen whenexercise was performed in a fed state.
Carbohydrates and fats are the most important sources of fuelduring rest and exercise(4). In general, the lipolytic activity ofadipose tissue is regulated by the balance between stimulatinghormones such as catecholamines and those that inhibit theenzyme responsible for lipolysis (lipase sensitive hormone),especially insulin(58). Because of higher muscular energyexigency and increased availability of NEFA(58), mediated byincreased adrenergic stimulation(59), exercise alone canincrease fat oxidation compared with rest(39).
Among the primary responses to fasting are the partial mobi-lisation of TAG reserves contained in the adipose tissue and thedecreased re-esterification of NEFA. This leads to an increase inthe concentration of circulating NEFA in plasma and, conse-quently, greater availability of this fuel source for the muscles(2,20).These fundamental principles can explain the findings of thepresent study, suggesting that when exercise is performed in thefasted state, lipolytic activity is increased further because of theaction of lipolysis-stimulating hormones and limited action ofinsulin. However, increased plasma concentrations of NEFAduring exercise are attenuated by carbohydrate intake beforeexercise, due to the inhibition of lipolysis mediated by insulin(6). Ithas also been suggested that increases in insulin concentrationscan directly inhibit the transfer of fat through the muscle cellmembrane and/or mitochondrial membranes(8). Therefore, as aconsequence of lower availability of NEFA and the inhibition ofoxidation of IMTG, exercise performed in the fed state showsreduced fat oxidation(7).
Apart from diet, use of energy substrates during exercisedepends on factors such as intensity, duration and level oftraining(4). It has been shown that fat oxidation, rather than theuse of carbohydrate as a substrate, tends to be higher at low-to-moderate intensities of exercise, no >60–65% VO2max, but islikely to decrease at an intensity >75% VO2max
(60,61). These datacorroborate the findings of the present study, in which most ofthe interventions relating to fat oxidation during exercise
Aziz et al., 2010
Bergman & Brooks, 1999a
Bergman & Brooks, 1999b
Bouhlel et al., 2006
Coyle et al., 1985
Coyle et al., 1997
Dohm et al., 1986
Farah & Gill., 2013
Gonalez et al., 2013
Guéye et al., 2003
Horowitz et al., 1997
Isacco et al., 2012a
Isacco et al., 2012b
Kirwan et al., 2001a
Kirwan et al., 2001b
Little et al., 2009
Little et al., 2010
Massicotte et al., 1990
Maughan & Gleeson, 1988
Montain et al., 1991a
Montain et al., 1991b
Paul et al., 1996
Ramos-Jiménez et al., 2014
Satabin et al., 1987
Schabort et al., 1999
Shin et al., 2013
Whitley et al., 1998
Willcutts et al., 1988
Wu et al., 2003
Ziogas & Thomas, 1998
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(pe
rfor
man
ce b
ias)
Blin
ding
of o
utco
me
asse
ssm
ent (
dete
ctio
n bi
as)
Inco
mpl
ete
outc
ome
data
(at
triti
on b
ias)
Fig. 3. Summary of risk of bias in the included studies.
Exercise in fasted adults and fat oxidation 1159
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
included in the meta-analysis were performed with intensitiesbetween 40 and 65% VO2max
(30,31,39,41,44,47,51,52). Studies thatused greater intensities, and where exact values were givenin the published report, did not surpass 70% VO2max
(40,43).Concerning the duration of exercise, the studies included wereevaluated up to 120min. It is suggested that after 2 h of exer-cising, the substrate utilisation patterns become similar betweenfasted and fed states(45). That is, an increase in fat oxidation mayalso occur in ‘fed state’ individuals after a certain interval, andmay be caused by a reduction in muscle glycogen that occurs inthe advanced stages of prolonged exercise(4). Furthermore, themajority of included studies were conducted with physicallyactive individuals(30,40,41,43,44,47,50–52), and the literature
indicates that fat oxidation during sub-maximal exercise isimproved with aerobic physical training(14,60).
As exercise intensity can influence the utilisation ofenergy substrates during exercise, sensitivity analyses wereperformed according to this criterion. It is well established inthe literature that the contribution of carbohydrate to energysupply increases incrementally with exercise intensity (>65%VO2max), whereas the fat oxidation peak occurs at lowerintensities (45–65% VO2max), which may be influenced by sex,training status, VO2max and diet(60). The present analysisshowed that during exercises at intensities <70% VO2max, fatoxidation was higher in the fasted state (approximately 3·45 g),but that there was no difference in this variable between
Bergman & Brooks, 1999aBergman & Brooks, 1999bFarah & Gill, 2013Gonzalez et al., 2013Isacco et al., 2012aIsacco et al., 2012bLittle et al., 2010Massicotte et al., 1990Maughan & Gleeson, 1988Paul et al., 1996Whitley et al., 1998Willcutts et al., 1988Wu et al., 2003
Difference in means and 95 % CIStatistics for each studyStudy name
Fig. 4. Fat oxidation (g) during exercise performed in the fasted state v. fed state. , Study-specific estimates: , pooled estimates of fixed-effects meta-analyses.
Favours fasted Favours fed
Coyle et aI., 1985Coyle et aI., 1997Dohm et aI., 1986Gonzalez et aI., 2013Horowitz et aI., 1997Isacco et aI., 2012aIsacco et aI., 2012bKirwan et aI., 2001aKirwan et aI., 2001bLittle et aI., 2010Massicotte et aI., 1990Paul et aI., 1996Satabin et aI., 1987Schabort et aI., 1999Shin et aI., 2013Whitley et aI., 1998Wu et aI., 2003
Difference in means and 95 % CIStatistics for each studyStudy name
Fig. 5. Weighted mean difference of NEFA concentrations (mmol/l) relative to exercise performed in the fasted state v. fed state. , Study-specific estimates:, pooled estimates of random-effects meta-analyses.
1160 A. F. Vieira et al.
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
fasted and fed states during exercises at intensities ≥70%VO2max. These findings confirm the results reported byBergman & Brooks(30), in which the effect of intensity andprevious feeding on energy substrate used during exercisewas verified. Higher fat oxidation was observed in the fastedstate, compared with the fed state, during exercises with
intensities up to 59% VO2peak, but not at the intensity of75% VO2peak.
The literature reports that physical training is able to reduceinsulin resistance(14) related to excessive accumulation of IMTGin sedentary individuals(62). This effect seems to be due toincreased fat oxidation(15), mainly coming from fatty acids
Differencein means SE Variance
Lowerlimit
Upperlimit Z-value P
Favours fasted Favours fed
Difference in means and 95 % CIStatistics for each studyStudy name
–4.00 –2.00 0.00 2.00 4.00
Aziz et al., 2010Bouhlel et al., 2006Coyle et al., 1985Coyle et al., 1997Dohm et al., 1986Farah & Gill, 2013Gonzalez et al., 2013Guéye et al., 2003Horowitz et al., 1997Isacco et al., 2012aIsacco et al., 2012bKirwan et al., 2001aKirwan et al., 2001bLittle et al., 2009Little et al., 2010Massicotte et al., 1990Montain et al., 1991aMontain et al., 1991bRamos-Jiménez et al., 2014Satabin et al., 1987Schabort et al., 1999Shin et al., 2013Wu et al., 2003Ziogas & Thomas, 1998
Fig. 6. Weighted mean difference of glucose concentrations (mmol/l) relative to exercise performed in the fasted state v. fed state. , Study-specific estimates:, pooled estimates of random-effects meta-analyses.
Study name Statistics for each study Difference in means and 95 % CI
SE
Lower Upperin meansDifference
Variance limit limit Z-value P
Coyle et al., 1985Coyle et al., 1997Dohm et al., 1986Gonzalez et al., 2013Horowitz et al., 1997Isacco et al., 2012aIsacco et al., 2012bKirwan et al., 2001aKirwan et al., 2001bLittle et al., 2010Massicotte et al., 1990Montain et al., 1991aMontain et al., 1991bSchabort et al., 1999Shin et al., 2013Whitley et al., 1998Wu et al., 2003
32.6 19.747.272.235.445.737.018.2
9.9125.613.017.323.649.64.4
40.118.742.817.2
200.4165.5107.5192.5103.1
923.715.645.182.6
188.2–6.9
196.434.2
188.9104.5
68.761.2
386.62232.1
1254.65218.9
2087.71372.0
330.997.9
15773.6170.1300.9555.6
2458.419.3
1610.5349.5
1833.2295.2
–5.9107.823.938.1
102.930.533.041.8
677.6–9.911.136.491.0
–15.5117.7–2.5
104.970.8
71.1293.0307.1176.9282.0175.7104.380.6
1169.941.279.1
128.8285.4
1.7275.170.8
272.8138.2
1.74.22.33.04.22.83.86.27.41.22.63.53.8
4.91.84.46.1
–1.6
0.0980.0000.0220.0020.000
0.0000.0000.000
0.0000.000
0.000
0.0000.000
–1200.00
0.005
0.231
0.117
0.068
0.009
–600.00 0.00 600.00 1200.00
Favours fasted Favours fed
Fig. 7. Weighted mean difference of insulin concentrations (pmol/l) relative to exercise performed in the fasted state v. fed state. , Study-specific estimates:, pooled estimates of random-effects meta-analyses.
Exercise in fasted adults and fat oxidation 1161
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
derived from IMTG(63). The acute effects of exercise in thefasted state are able to reduce the content of IMTG byapproximately 60%(5,64), which does not seem to occur in thefed state(5) and, in the long term, seems to be more effective inimproving insulin sensitivity(65).Venables & Jeukendrup(14) demonstrated that an increase in
fat oxidation of approximately 3 g during 30min of aerobicexercise was able to enhance insulin sensitivity in obese andsedentary men. The present meta-analysis indicates that aerobicexercise performed in the fasted state provides an increase in fatoxidation of about 3·08 g during the session compared with thefed state. Therefore, it is suggested that exercising in the fastedstate can be an alternative to increase the use of fat as theenergy source and the increase of oxidised fats by 3·08 g duringan exercise session may be sufficient to induce improvementsin insulin sensitivity.As previously described, it is well established that plasma
concentrations of NEFA are higher in the fasted states comparedwith fed states(4,66,67). However, our results indicate that themagnitude of variation, from before to after exercise, does notappear to differ between fasted v. fed states. In general, during thefirst 15min of exercise, plasma NEFA concentrations decrease, asthe utilisation rate in the muscles exceeds the lipolysis-drivenrelease rate. After this period, the release rate exceeds use in themuscles and the fatty acid concentrations in the plasma rise(4). Onthe basis of the results provided in this meta-analysis, this eventseems to occur in similar ways in both fasted and fed states.Although the weighted mean difference of NEFA showed no
significant differences, the present study demonstrates that varia-tion in glucose and insulin concentrations before and after exer-cise was significantly higher during exercise performed in the fedstate. One possible explanation for this finding in relation toinsulin is that carbohydrate ingestion before exercise can result ina considerable increase in insulin concentrations(68), which mayremain high for about 3 h after consuming a meal(51), and tend toreturn to basal values when exercise is performed(45,47). In thiscase, it is noteworthy that the majority of studies included in themeta-analysis offered meals up to 180min before exercis-ing(6,7,31,32,37,38,47–52). Hence, it is probable that insulin concentra-tions remained high at the beginning of exercise and decreasedover the course of the exercise for ‘fed state’ participants.Regarding glucose concentrations, the highest variation
generated by exercise performed in the fed state is attributed toincreased glucose concentrations in plasma, due to the intake ofcarbohydrates before exercise, and subsequent fall in glucoseconcentration due to the combined effects of hyperinsulinaemiaand glucose uptake for use as an energy substrate in musclecontractile activity(6,69). On the other hand, fasting causesincreases in glycerol release through hydrolysis of TAGmolecules from fat cells; this is a valuable precursor forhepatic gluconeogenesis, thus contributing to the availability ofglucose(2). These principles can be used to explain the greatervariations in plasma glucose concentration in the fed staterelative to the fasted state. When sensitivity analyses wereperformed for this variable according to the criterion‘quantity of carbohydrate consumed in the pre-exercise meal≥100 g’, meta-analysis did not demonstrate a significant differ-ence in the variation of glucose between fasted v. fed states.
A possible hypothesis for this is that the intake of carbohydrate-rich meals increases the availability of glucose duringexercise(68).
Although this systematic review with meta-analysis was per-formed with the maximum methodological rigour possible, somelimitations should be highlighted. First was the inclusion of dif-ferent intensities and durations of exercise, sex, times betweenmeals and exercise, types of meals and quality and quantity ofcarbohydrates: in spite of methodological differences between thestudies under review, we sought to maximise standardisation inthe data examined. Second, as many relatively old publicationswere included in this study, certain methodological limitationsand flaws were noted in the presentation of data. We would liketo emphasise in particular that the large number of trials withresults presented exclusively in graphs, and the lack of provisionof these data (means and standard deviations) from authors,limited the accuracy of data extraction. Third, most of the studiesincluded in the analyses of glucose concentrations assessed glu-cose levels using venous blood sampling. Previous studies havereported(70,71) that the arterial sampling would be more recom-mended to assess glucose levels; however, venous blood sam-pling is the most commonly used method and it is widelyaccepted. In addition, high heterogeneity was identified in meta-analyses related to blood molecular concentrations, necessitatingcaution in interpreting these data.
Is worth mentioning that, although our results have shownincreased fat oxidation during exercise performed in the fastedstate, it is necessary to take care when prescribing this strategyin practice, as this meta-analysis was performed using only dataassessing the acute effects of exercise during fasting v. fedstates. The findings should not be extrapolated as long-termeffects, especially with the aim of reducing body fat, as there isinsufficient evidence of effectiveness and safety.
Conclusion
This systematic review with meta-analysis suggests that aerobicexercise at low-to-moderate intensity, performed in the fastedstate, induces an increase in fat oxidation, when compared withexercise performed following consumption of a carbohydrate-containing meal. Despite high heterogeneity of the data, nodifference appears to exist between exercising in the fasted orfed states in relation to variations in NEFA concentrations beforeand after exercise. In contrast, variation in relation to glucoseand insulin concentrations appears to be higher in the fedstates. Future meta-analyses and randomised clinical trials,inclusive of an evaluation of the long-term effects of aerobicexercise on fat and carbohydrate metabolism in the fasted andfed states, will be necessary to confirm the findings of thepresent review, as well as to identify their real benefits orconsequences for long-term health.
Acknowledgements
The authors thank François Péronnet, Jonathan Little, JohnKirwan, Laurie Isacco, Nathalie Boisseau, Bryan Bergman,
1162 A. F. Vieira et al.
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
Karen Soo, Donald Chisholm and Katarina Borer, all of whomanswered our questions by email.The present study was supported by the following organi-
sations: CAPES, CNPq and FAURGS.The authors’ contributions are as follows: A. F. V., R. C. O. M.
and R. R. C. formulated the research questions; A. F. V., R. C. O. M.,R. R. C. and L. F. M. K. designed the study and A. F. V., R. R. C. andL. C. performed the study; A. F. V. and R. R. C. analysed the data;A. F. V., R. R. C. and R. C. O. M. wrote the paper. All the authorscritically reviewed and improved the manuscript.The authors declare that there are no conflicts of interest.
Supplementary material
For supplementary material/s referred to in this article, pleasevisit http://dx.doi.org/doi:10.1017/S0007114516003160
References
1. Longo VD & Mattson MP (2014) Fasting: molecular mechan-isms and clinical applications. Cell Metab 19, 181–192.
2. Maughan RJ, Fallah J & Coyle EF (2010) The effects offasting on metabolism and performance. Br J Sports Med 44,490–494.
3. Cahill GF Jr (2006) Fuel metabolism in starvation. Annu RevNutr 26, 1–22.
4. Jeukendrup AE (2003) Modulation of carbohydrate and fatutilization by diet, exercise and environment. Biochem SocTrans 31, 1270–1273.
5. De Bock K, Richter EA, Russell AP, et al. (2005) Exercise in thefasted state facilitates fibre type-specific intramyocellular lipidbreakdown and stimulates glycogen resynthesis in humans.J Physiol 564, 649–660.
6. Horowitz JF, Mora-Rodriguez R, Byerley LO, et al. (1997)Lipolytic suppression following carbohydrate ingestion limitsfat oxidation during exercise. Am J Physiol 273, E768–E775.
7. Coyle EF, Jeukendrup AE, Wagenmakers AJ, et al. (1997) Fattyacid oxidation is directly regulated by carbohydrate metabo-lism during exercise. Am J Physiol 273, E268–E275.
8. Spriet LL (2014) New insights into the interaction of carbo-hydrate and fat metabolism during exercise. Sports Med 44,S87–S96.
9. Bassuk SS & Manson JE (2005) Epidemiological evidence forthe role of physical activity in reducing risk of type 2 diabetesand cardiovascular disease. J Appl Physiol 99, 1193–1204.
10. Donnelly JE, Blair SN, Jakicic JM, et al. (2009) American Col-lege of Sports Medicine Position Stand. Appropriate physicalactivity intervention strategies for weight loss and preventionof weight regain for adults. Med Sci Sports Exerc 41, 459–471.
11. Slentz CA, Duscha BD, Johnson JL, et al. (2004) Effects of theamount of exercise on body weight, body composition, andmeasures of central obesity. Arch Intern Med 164, 31–39.
12. Goodpaster BH, Katsiaras A & Kelley DE (2003) Enhanced fatoxidation through physical activity is associated withimprovements in insulin sensitivity in obesity. Diabetes 52,2191–2197.
13. Rosenkilde M, Nordby P, Nielsen LB, et al. (2010) Fat oxidationat rest predicts peak fat oxidation during exercise and meta-bolic phenotype in overweight men. Int J Obes 34, 871–877.
14. Venables MC & Jeukendrup AE (2008) Endurance training andobesity: effect on substrate metabolism and insulin sensitivity.Med Sci Sports Exerc 40, 495–502.
15. Robinson SL, Hattersley J, Frost GS, et al. (2015) Maximal fatoxidation during exercise is positively associated with 24-hourfat oxidation and insulin sensitivity in young, healthy men.J Appl Physiol 118, 1415–1422.
16. Burton FL, Malkova D, Caslake MJ, et al. (2010) Substratemetabolism, appetite and feeding behavior under low andhigh energy turnover conditions in overweight women.Br J Nutr 104, 1249–1259.
17. Iwayama K, Kawabuchi R, Park I, et al. (2015a) Transientenergy deficit induced by exercise increases 24-h fat oxidationin young trained men. J Appl Physiol 118, 80–85.
18. Iwayama K, Kurihara R, Nabekura Y, et al. (2015b) Exerciseincreases 24-h fat oxidation only when it is performed beforebreakfast. EBioMedicine 2, 2003–2009.
19. Shimada K, Yamamoto Y, Iwayama K, et al. (2013) Effects ofpost-absorptive and postprandial exercise on 24 h fat oxida-tion. Metabolism 62, 793–800.
20. Enevoldsen LH, Simonsen L, Macdonald IA, et al. (2004) Thecombined effects of exercise and food intake on adiposetissue and splanchnic metabolism. J Physiol 561, 871–882.
21. Lee BM & Wolever TMS (1998) Effect of glucose, sucrose andfructose on plasma glucose and insulin responses in normalhumans: comparison with white bread. Eur J Clin Nutr 52,924–928.
22. Liberati A, Altman DG, Tetzlaff J, et al. (2009) The PRISMAstatement for reporting systematic reviews and meta-analysesof studies that evaluate health care interventions: explanationand elaboration. Ann Intern Med 151, W-65–W-94.
23. Shamseer L, Moher D, Clarke M, et al. (2015) PreferredReporting Items for Systematic Review and Meta-AnalysisProtocols (PRISMA-P) 2015: elaboration and explanation.BMJ 349, 1–25.
24. Péronnet F & Massicotte D (1991) Table of nonproteinrespiratory quotient: an update. Can J Sport Sci 16, 23–29.
25. Jeukendrup AE & Wallis GA (2005) Measurement of substrateoxidation during exercise by means of gas exchangemeasurements. Int J Sports Med 26, S28–S37.
26. Deeks JJ, Higgins JPT & Altman DG (editors) (2011) Chapter 9:Analysing data and undertaking meta-analysis. In CochraneHandbook for Systematic Reviews of Interventions, Version5.1.0 (updated March 2011) [JPT Higgins and S Green, editors].The Cochrane Collaboration. http://handbook.cochrane.org/CCAa (accessed September 2015).
27. Higgins JPT, Deeks JJ & Altman DG (editors) (2008) Chapter 16:Special topics in statistics. In Cochrane Handbook for SystematicReviews of Interventions, Version 5.0.1 (updated September2008) [JPT Higgins and S Green, editors]. The CochraneCollaboration. http://www.cochrane-handbook.org (accessedJuly 2015).
28. Higgins JPT, Thompson SG, Deeks JJ, et al. (2003) Measuringinconsistency in meta-analyses. BMJ 327, 557–560.
29. Egger M, Davey Smith G, Schneider M, et al. (1997) Bias inmeta-analysis detected by a simple, graphical test. BMJ 315,629–634.
30. Bergman BC & Brooks GA (1999a, b) Respiratory gas-exchange ratios during graded exercise in fed and fastedtrained and untrained men. J Appl Physiol 86, 479–487.
31. Isacco L, Thivel D, Pelle AM, et al. (2012a, b) Oral contraceptionand energy intake in women: impact on substrate oxidationduring exercise. Appl Physiol Nutr Metab 37, 646–656.
32. Montain SJ, Hopper MK, Coggan AR, et al. (1991a, b) Exercisemetabolism at different time intervals after a meal. J ApplPhysiol 70, 882–888.
33. Drenick EJ, Fisler JS, Johnson DG, et al. (1977) The effect ofexercise on substrates and hormones during prolongedfasting. Int J Obes 1, 49–61.
Exercise in fasted adults and fat oxidation 1163
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
34. Borer KT, Wuorinen E, Chao C, et al. (2005) Exercise energyexpenditure is not consciously detected due to oro-gastric, notmetabolic, basis of hunger sensation. Appetite 45, 177–181.
35. Erdmann J, Tholl S & Schusdziarra V (2010) Effect of carbohy-drate- and protein-rich meals on exercise-induced activation oflipolysis in obese subjects. Horm Metab Res 42, 290–294.
36. Little JP, Chilibeck PD, Ciona D, et al. (2009) The effects oflow- and high-glycemic index foods on high-intensity inter-mittent exercise. Int J Sports Physiol Perform 4, 367–380.
37. Schabort EJ, Bosch AN, Weltan SM, et al. (1999) The effect of apreexercise meal on time to fatigue during prolonged cyclingexercise. Med Sci Sports Exerc 31, 464–471.
38. Shin YH, Jung HL, Ryu JW, et al. (2013) Effects of a pre-exercise meal on plasma growth hormone response and fatoxidation during walking. Prev Nutr Food Sci 18, 175–180.
39. Farah NMF & Gill JMR (2013) Effects of exercise before or aftermeal ingestion on fat balance and postprandial metabolism inoverweight men. Br J Nutr 109, 2297–2307.
40. Maughan RJ & Gleeson M (1988) Influence of a 36h fastfollowed by refeeding with glucose, glycerol or placebo onmetabolism and performance during prolonged exercise inman. Eur J Appl Physiol 57, 570–576.
41. Paul GL, Rokusek JT, Dykstra GL, et al. (1996) Oat, wheat orcorn cereal ingestion before exercise alters metabolismin humans. J Nutr 126, 1372–1381.
42. Satabin P, Portero P, Defer G, et al. (1987) Metabolic andhormonal responses to lipid and carbohydrate diets duringexercise in man. Med Sci Sports Exerc 19, 218–223.
43. Whitley HA, Humphreys SM, Campbell IT, et al. (1998)Metabolic and performance responses during enduranceexercise after high-fat and high-carbohydrate meals. J ApplPhysiol 85, 418–424.
44. Willcutts KF, Wilcox AR & Grunewald KK (1988) Energymetabolism during exercise at different time intervalsfollowing a meal. Int J Sports Med 9, 240–243.
45. Coyle EF, Coggan AR, Hemmert MK, et al. (1985) Substrateusage during prolonged exercise following a preexercisemeal. J Appl Physiol 59, 429–433.
46. DohmGL, Beeker RT, Israel RG, et al. (1986) Metabolic responsesto exercise after fasting. J Appl Physiol 61, 1363–1368.
47. Gonzalez JT, Veasey RC, Rumbold PLS, et al. (2013) Breakfast andexercise contingently affect postprandial metabolism and energybalance in physically active males. Br J Nutr 110, 721–732.
48. Kirwan JP, Cyr-Campbell D, Campbell WW, et al. (2001a)Effects of moderate and high glycemic index meals on meta-bolism and exercise performance. Metabolism 50, 849–855.
49. Kirwan JP, O’Gorman DJ, Cyr-Campbell D, et al. (2001b)Effects of a moderate glycemic meal on exercise duration andsubstrate utilization. Med Sci Sports Exerc 33, 1517–1523.
50. Little JP, Chilibeck PD, Ciona D, et al. (2010) Effect of low andhigh glycemic-index meals on metabolism and performanceduring high-intensity, intermittent exercise. Int J Sport NutrExerc Metab 20, 447–456.
51. Wu CL, Nicholas C, Williams C, et al. (2003) The influence ofhigh-carbohydrate meals with different glycaemic indices onsubstrate utilisation during subsequent exercise. Br J Nutr 90,1049–1056.
52. Massicotte D, Péronnet F, Brisson G, et al. (1990) Oxidation ofexogenous carbohydrate during prolonged exercise in fedand fasted conditions. Int J Sports Med 11, 253–258.
53. Aziz AR, Wahid MF, Png W, et al. (2010) Effects ofRamadan fasting on 60 min of endurance running perfor-mance in moderately trained men. Br J Sports Med 44,516–521.
54. Bouhlel E, Salhi Z, Bouhlel H, et al. (2006) Effect of Ramadanfasting on fuel oxidation during exercise in trained male rugbyplayers. Diabetes Metab 32, 617–624.
55. Guéye L, Seck D, Samb A, et al. (2003) Physiological adap-tations to exercise during a short-term fasting. Scr Med (Brno)76, 291–296.
56. Ramos-Jiménez A, Anguiano-Juarez J, Sifuentes-Juarez A, et al.(2014) Fasting and postprandial glycemia in response to astrenuous workout in healthy subjects with family history ofdiabetes and borderline insulin resistance. Exp Clin Cardiol20, 139–161.
57. Ziogas G & Thomas TR (1998) Dietary preparation before restand exercise testing. Nutrition 14, 11–16.
58. Horowitz JF & Klein S (2000) Lipid metabolism duringendurance exercise. Am J Clin Nutr 72, 558S–563S.
59. Arner P, Kriegholm E, Engfeldt P, et al. (1990) Adrenergicregulation of lipolysis in situ at rest and during exercise. J ClinInvest 85, 893–898.
61. van Loon LJ, Greenhaff PL, Constantin-Teodosiu D, et al.(2001) The effects of increasing exercise intensity on musclefuel utilization in humans. J Physiol 536, 295–304.
62. Kelley DE & Goodpaster BH (2001) Skeletal muscle trigly-ceride. An aspect of regional adiposity and insulin resistance.Diabetes Care 24, 933–941.
63. Schrauwen P, van Aggel-Leijssen DP, Hul G, et al. (2002) Theeffect of a 3-month low-intensity endurance training programon fat oxidation and acetyl-CoA carboxylase-2 expression.Diabetes 51, 2220–2226.
64. van Loon LJC, Koopman R, Stegen JHCH, et al. (2003) Intra-myocellular lipids form an important substrate source duringmoderate intensity exercise in endurance-trained males in afasted state. J Physiol 553, 611–625.
65. Van Proeyen K, Szlufcik K, Nielens H, et al. (2010) Training inthe fasted state improves glucose tolerance during fat-rich diet. J Physiol 588, 4289–4302.
66. Sahlin K & Harris RC (2008) Control of lipid oxidation duringexercise: role of energy state and mitochondrial factors.Acta Physiol (Oxf) 194, 283–291.
67. Spriet LL & Watt MJ (2003) Regulatory mechanisms in theinteraction between carbohydrate and lipid oxidation duringexercise. Acta Physiol Scand 178, 443–452.
68. Hargreaves M, Hawley JA & Jeukendrup A (2004) Pre-exercisecarbohydrate and fat ingestion: effects on metabolism andperformance. J Sports Sci 22, 31–38.
69. Marmy-Conus N, Fabris S, Proietto J, et al. (1996) Pre-exerciseglucose ingestion and glucose kinetics during exercise. J ApplPhysiol 81, 853–857.
70. Liu D, Moberg E, Kollind M, et al. (1992) Arterial,arterialized venous, venous and capillary blood glucosemeasurements in normal man during hyperinsulinaemiceuglycaemia and hypoglycaemia. Diabetologia 35,287–290.
71. Brouns F, Bjorck I, Frayn KN, et al. (2005) Glycaemic indexmethodology. Nutr Res Rev 18, 145–171.
1164 A. F. Vieira et al.
Dow
nloaded from https://w
ww
.cambridge.org/core . IP address: 54.39.106.173 , on 23 M
ay 2020 at 01:01:51 , subject to the Cambridge Core term
